Cross-scale dynamics and the evolutionary emergence of infectious diseases

Sebastian J. Schreiber; Ruian Ke; Claude Loverdo; Miran Park; Priyanna Ahsan; James O. Lloyd-Smith

In this model, the viral population initially increases exponentially because of the availability of a large number of target cells. The wild-type and mutant strains increase exponentially at rates, r w and r m , respectively. These dynamics are consistent with the viral dynamics predicted by standard multistrain within-host models when target cells are abundant [86] [87] [88] . The viral load saturates at time T e with a maximal viral load K (Fig 1B) . We assume that after T e , the viral population size stays constant at K, and the frequencies of the wild-type and the mutant change due to their fitness differences. We further assume that the infectious period starts when v w (t) + v m (t) > 0 and ends after T days. For some acute infections, viral load usually decreases rapidly after viral peak (e.g., influenza A infections [89] ), and thus T is close to T e . For other acute infections, such as SARS-CoV-2 [90] infection, viral load remains at a high level after peak viremia for a longer period of time, in the range of weeks. In this case, T > T e . For chronic infections (though our work does not necessarily consider the full range of evolutionary processes involved in chronic infections) such as HIV and Hepatitis C [91] infections, viral loads usually reach a set-point and can stay roughly constant for an extended period of time, in the range of years. In this case, T is much greater than T e . Overall, this within-host model serves as a flexible framework to describe a range of viral dynamics from both acute and chronic infections, while maintaining simplicity to enable analysis.